Here is a great article explaining the virtues of having some infrastructure assets in your portfolio:
16 October 2008 | by Dennis Eager
The threat of global inflation is now once again on the agenda, and it is time for investors to rethink their portfolios and consider the impact of inflation.
Infrastructure is a new asset class to many investors. It is increasingly seen as an attractive addition to a portfolio due to its inflation-linked revenues, low operating costs and high EBITDA (earnings before interest, taxes, depreciation and amortisation) margins. Additionally, the relatively high levels of debt carried by infrastructure assets are prudently managed and often hedged, muting exposure to interest rate movements. These combined characteristics provide an attractive platform for investors seeking a ‘safe haven’ in an inflationary environment.
A sustained increase in inflation will negatively impact the value of any company. From a valuation perspective, nominal bond yields will increase, increasing the risk-free rate of return (in practice, the yield on a long-term Government bond) and therefore the cost of equity, reducing the valuation of the company. From a fundamental perspective, input costs will increase and it is likely that higher short-term interest rates will increase borrowing costs.
To the extent that any company is unable to pass through increases in input costs or borrowing costs without impacting demand or market share, inflation will negatively impact profits. The impact will be determined by the buying power, competitive position, and leverage of the company, as well as the price elasticity of demand for the company’s products or services. Very few companies operating in competitive markets will not suffer some diminution in value if inflation rises.
Impact of inflation on infrastructure
This paper adopts a strict definition of infrastructure under which two criteria must be fulfilled:
i. the asset either has to be a monopoly or behave like a monopoly; and
ii. the asset must be required for the efficient functioning of a community.
This definition is particularly important as assets that meet these requirements provide significant cash flow predictability and, therefore, an attractive investment proposition within the defensive component of an investor’s portfolio. The primary sectors falling within this definition are utilities, toll roads, airports, ports, and communications. Each exhibits different investment characteristics and is impacted differently by increases in inflation.
Utilities include electricity transmission (high voltage power lines) and electricity distribution (urban power lines) as well as gas transmission and distribution.
In most markets (unlike Australia), these are pure monopolies. Prices these entities charge to transport the electricity or gas are controlled by a government-appointed regulator, which, after a period of public consultation, will adjust charge-out rates to provide the utility with an appropriate return on invested capital. The timing of these regulatory resets varies according to the jurisdiction. Generally, reviews are annual or at the request of either the utility or community groups. The utility is therefore afforded a high level of protection against inflation.
Australia is the exception, with reviews completed every five years, leaving the assets more exposed to inflation in the short term. There are four major types of participants in the Australian electricity industry:
- Energy generators using coal or gas to generate electricity. There are numerous generating firms and this segment is competitive;
- Electricity is then transported from generators over high voltage transmission lines by transmission companies. There is no need for multiple transmission lines, so this segment is a natural monopoly and is regulated by the Australian Energy Regulator;
- Distribution companies transport the electricity over low voltage lines to consumers. As with transmission, it is not practical to build multiple sets of distribution and this segment is also a natural monopoly and is regulated by state-based regulators; and
- Electricity retailers arrange for the delivery of electricity to the consumer. There are a number of companies that undertake this role and compete directly and, hence, this function is not regulated.
Similarly, in the Australian gas industry, producers are not regulated, as transmission pipelines are regulated by the Australian Energy Regulator, distribution pipelines transporting gas to the end user are regulated by state-based regulators, and retailers are not regulated.
Finally, utilities exhibit low price elasticity of demand. Very high barriers to entry (due to the natural monopoly characteristics of power lines and gas pipelines) prevent competition. Coupled with these entry barriers, there is usually no available alternative for the electricity or gas producer to transport their product to the end-user, therefore demand is highly price inelastic.
Around the world, the typical business model for a toll road sees a government agency entering into a concession agreement (contract) that entitles a company to collect tolls for a defined period and increase those tolls on a regular basis in a defined way. At the end of this concession, the road is handed back to the government in a good state of repair.
In most markets, the toll road is not the only road route available to motorists (water crossings such as bridges being an exception). Consequently, the toll road is not a pure monopoly. However, the toll road is generally built in the first place because either the alternative routes are not suitable for high speed distance travel – for example, almost all European toll roads offer the only viable route for long-distance travellers with time imperatives – or the alternative routes are highly congested.
The opening of a new toll road inevitably reduces traffic on the free alternative. But over time, the free alternative will become heavily congested much more quickly than the toll road. As that occurs, the toll road behaves much like a monopoly. Table 1 shows the demand impact of toll increases on two Sydney toll roads over the last decade. Traffic has been very price inelastic, except for the first toll increase on the M5 back in 1996.
As mentioned, the basis on which tolls are increased is controlled by the terms of the concession agreement. There are only three toll roads of any significance in the Western world where the concessions effectively allow the concessionaire the discretion on toll increases (the M6 toll in the UK, the 407ETR in Canada and the SR125 in California). All other concessions have specific formulae that are generally related to inflation. Table 2 provides a typical cross section.
As can be seen, the pricing mechanism for these toll roads picks up any increases in inflation with minimal lag. Consequently, the majority of toll roads have the ability to respond quickly to any spike in inflation. And as the data in Table 1 highlights, there will be minimal if any loss in traffic when tolls increase, so revenues will fully recover the inflationary impact.
Another key characteristic of toll roads that insulates them from inflationary impacts is their high EBITDA margins. Table 3 shows the EBITDA margin for a selection of international toll roads. The average EBITDA margin is 77 per cent, substantially above the 14.6 per cent average EBITDA margin of other industrial companies globally (Merrill Lynch 2008).
Inflation can also have a significant impact on capital expenditure. With most toll roads, capital expenditure on operating roads is minimal and generally limited to resurfacing and replacement of ageing crash barriers, etcetera. For example, Hills Motorway, the entity that owns the M2 Motorway in Sydney, generated $123 million in revenue in the 2007 financial year, with $4.3 million of capital expenditure, less than 4 per cent of revenues. These economics are typical of toll roads and thus inflation has no material importance to capital expenditure for the sector.
Finally, like most infrastructure assets, toll roads are generally more highly geared than average industrial companies. This is discussed further below.
Airports should be considered as two separate businesses – airside and landside operations.
Airside operations primarily involve the management of the runways and taxiways of the airport. Airside revenue is generated by either a charge levied per passenger or a charge levied on the weight of the plane or a combination of both. In most jurisdictions, the onus is on the airport to negotiate appropriate charges with the airlines, with some form of regulation as a fall back position. This side of the operations therefore behaves much like a regulated utility.
Landside operations involve the remainder of the airport and fall into three primary areas: retail, car parking and property development. Generally, the airport does not directly run retail outlets. Instead, it acts as the lessor and receives a guaranteed minimal rental (normally consumer price index (CPI) linked) and a share of sales. There is therefore a direct protection of these revenues from a spike in inflation. Car parking operations generally behave like a monopoly, although there is some substitution threat (e.g, the potential for airport users to use a taxi instead of driving). Still, the airport has significant potential to increase prices in response to inflationary spikes.
Airport operating margins exhibit much greater variability than toll roads, as evident in Table 4.
The more efficient airports like Sydney and Auckland are better insulated from inflationary spikes than those (typically European) airports that are still struggling to reduce the workforces that were in place when they were privatised (notably, these less efficient airports still exhibit higher EBITDA margins than the average industrial company).
Finally, airports also have the highest capital expenditure requirements of any infrastructure sub sector. Capital expenditure includes the widening and extension of runways and taxiways and is generally only undertaken after consultation and agreement with the airlines and any regulatory authority. Consequently, airside charges will be increased to recover these costs over time. Landside capital expenditure relates to increasing retail, parking and general property-leasing facilities. Higher inflation may impact the financial viability of such capital expenditure but the airport has an unregulated monopoly in these areas and can increase prices as required to recover inflation. Consequently, inflation is unlikely to have a material impact on the value of an airport asset as a result of higher capital expenditure.
Ports and communications
Ports and broadcast communication towers make up less than 5 per cent of the infrastructure sector. Ports face economics very similar to those exhibited by the airside operations of airports, however they typically face lower levels of regulation (or regulatory threat) and thus exhibit real price increases over time and relatively high EBITDA margins. Communication assets relate to the carriage and delivery of information (such as television and radio broadcasts, telephones and data services). These face very little susceptibility to inflation due to pricing power driven by high barriers to entry and an inelastic demand profile.
Many infrastructure companies are able to obtain relatively cheap and long-term debt by comparison with the average industrial company. Gearing levels are generally high by industrial company standards – for example, utilities carry a net debt/equity ratio of 78.1 per cent (transport assets are similar) while the average industrial company carries a net debt/equity ratio of 48.4 per cent. Hence, management of debt exposures is of significant importance to infrastructure companies and generally very professionally managed. During the turmoil in global credit markets, very few infrastructure stocks globally have experienced any problems in meeting debt obligations or funding new debt requirements.
As already discussed, regulated utilities have the ability to recover the cost implications of any inflationary spike through the periodic regulatory process. This generally includes the costs of servicing higher interest rates on debt, therefore exposure to interest rates will be limited to the length of time between reset periods.
But the other infrastructure sectors do not enjoy such automatic linkages – as demonstrated above, their revenue streams enjoy inflation linkages, however there are few mechanisms to recover increased borrowing costs from customers. In light of this, the management of most infrastructure companies normally swap floating rate debt obligations for fixed rate debt, thereby eliminating (at least until the swaps terminate) any exposure to short-term interest rate spikes. Consequently, the majority of infrastructure companies are well protected from any inflationary spike.
The role of infrastructure in investor portfolios has been discussed in various publications and therefore this paper will not investigate quantitative portfolio considerations. For instance, papers published previously in the PortfolioConstruction Journal have argued an allocation of 8 per cent to 10 per cent to infrastructure is appropriate in a balanced portfolio.
The key question is how to fund an allocation to infrastructure. Some investors consider infrastructure as a potential replacement for property in a portfolio.
This should be treated with caution for two primary reasons (as cited above). Firstly, the inflation linkages of infrastructure listed above do not hold true for property, and rental income streams typically increase at a rate below CPI. Secondly, property and infrastructure exhibit robust revenue streams with a lower correlation to economic activity than other industrial companies – however, rental income streams derived from property investments are likely to have a higher correlation to economic activity. For both reasons, the risk/return outcomes from property and infrastructure should be expected to be different over time and therefore infrastructure should be considered a standalone asset class.
Dennis Eagar is portfolio manager at Magellan Asset Management. This paper is abridged from a full research paper presented at the seventh annual PortfolioConstruction Conference (August 27-28, 2008). The full paper is available